window_size

Features

HistogramSampling()

HistogramSampling(histogram, bins_midpoints, samples)

Function for doing histogram sampling from a distribution

Parameters:

histogram (array) – histogram counts normalised to get a sum =1

bins_midpointsarray

values corresponding to the probability values given as “histogram”

samplesint

number of histograms samples

returns:

• hist2 (array) – histogram counts of the resulting distribution

• bins_midpoints (array) – midpoints of bins corresponding to hist2

References

• Marwan, N., Romano, M. C., Thiel, M., & Kurths, J. (2007). Recurrence plots for the analysis of complex systems. Physics reports, 438 (5-6), 237–329.

Sliding_windowBS_sub_per_lam()

Sliding_windowBS_sub_per_lam(RP, maxsize, winsize, n_boot)

Function for computing the CI of the bootstrapped confidence interval for percentage laminarity

Parameters:

• RP (ndarray) – recurrence plot

• maxsize (int) – RP size

• winsize (int) – size of the sliding window

  n_bootint

  number of bootstrap samples

Returns:

CI (double) – confidence interval(95% quantile - 5% quantile)

References

• Marwan, N., Romano, M. C., Thiel, M., & Kurths, J. (2007). Recurrence plots for the analysis of complex systems. Physics reports, 438 (5-6), 237–329.

Sliding_windowBS_sub_avg_vert()

Sliding_windowBS_sub_avg_vert(RP, maxsize, winsize, n_boot)

Function for computing the CI of the bootstrapped confidence interval for average vertical line length

Parameters:

• RP (ndarray) – recurrence plot

• maxsize (int) – RP size

• winsize (int) – size of the sliding window

  n_bootint

  number of bootstrap samples

Returns:

CI (double) – confidence interval(95% quantile - 5% quantile)

References

• Marwan, N., Romano, M. C., Thiel, M., & Kurths, J. (2007). Recurrence plots for the analysis of complex systems. Physics reports, 438 (5-6), 237–329.

Sliding_windowBS_sub_percent_det()

Sliding_windowBS_sub_percent_det(RP, maxsize, winsize, n_boot)

Function for computing the CI of the bootstrapped confidence interval for percentage determinism

Parameters:

• RP (ndarray) – recurrence plot

• maxsize (int) – RP size

• winsize (int) – size of the sliding window

  n_bootint

  number of bootstrap samples

Returns:

CI (double) – confidence interval(95% quantile - 5% quantile)

References

• Marwan, N., Romano, M. C., Thiel, M., & Kurths, J. (2007). Recurrence plots for the analysis of complex systems. Physics reports, 438 (5-6), 237–329.

Sliding_windowBS_sub_avg_diag()

Sliding_windowBS_sub_avg_diag(RP, maxsize, winsize, n_boot)

Function for computing the CI of the bootstrapped confidence interval for average diagonal line length

Parameters:

• RP (ndarray) – recurrence plot

• maxsize (int) – RP size

• winsize (int) – size of the sliding window

  n_bootint

  number of bootstrap samples

Returns:

CI (double) – confidence interval(95% quantile - 5% quantile)

References

• Marwan, N., Romano, M. C., Thiel, M., & Kurths, J. (2007). Recurrence plots for the analysis of complex systems. Physics reports, 438 (5-6), 237–329.

Sliding_windowBS()

Sliding_windowBS(RP, maxsize, var, n_boot=1000)

Function for computing the CI of the bootstrapped confidence interval for a given variable for different window sizes starting from 20 to a maximum of RP size incremented by value of 10

Parameters:

• RP (ndarray) – recurrence plot

• maxsize (int) – RP size

• var (str) – RQA variable name

  percent_lam : percentage laminarity

  percent_det : percentage determinism

  avg_vert : average value of vertical line distribution

  avg_diag : average value of diagonal line distribution

  n_bootint

  number of bootstrap samples

Returns:

data (dataframe) – dataframe containing the CI estimate of corresponding window sizes, columns are

WINSIZE : window size

95% quantile- 5% quantile : confidence interval

References

• Marwan, N., Romano, M. C., Thiel, M., & Kurths, J. (2007). Recurrence plots for the analysis of complex systems. Physics reports, 438 (5-6), 237–329.

Sliding_window_whole_data()

Sliding_window_whole_data(RP_dir, var, n_boot=1000)

Function for computing the CI of the bootstrapped confidence interval for a given variable for different window sizes starting from 20 to a maximum of RP size incremented by value of 10. Done for all datasets in the folder specified

Parameters:

• RP_dir (str) – directory containing all the RP files(.npy)

• var (str) – RQA variable name

  percent_lam : percentage laminarity

  percent_det : percentage determinism

  avg_vert : average value of vertical line distribution

  avg_diag : average value of diagonal line distribution

  n_bootint

  number of bootstrap samples

Returns:

data (dataframe) – dataframe containing the CI estimate of corresponding window sizes, columns are

WINSIZE : window size

95% quantile- 5% quantile : confidence interval

group : filename of the RPs

References

• Marwan, N., Romano, M. C., Thiel, M., & Kurths, J. (2007). Recurrence plots for the analysis of complex systems. Physics reports, 438 (5-6), 237–329.